This analysis examines the relative strength of Obama and Romney over the past nine months, using all available state head-to-head polls (something over 200 of them).
There haven’t been any new state head-to-head polls released in this race since just before Christmas. The pollsters went on vacation, and have since turned their attention to primary polls.
I’ve used the week since the previous analysis to hunt down older polls for this race going back to late 2010. I’ve also double checked the numbers from my first flurry of entering poll data into the computer and found two errors: I flipped the Obama and Romney numbers in a PA poll (which now causes the state to look a little bluer), and I fixed an incorrect sample size in a Georga poll. Among the newly-discovered (but older) polls, I’ve found the only poll from ND and a TN poll that is more recent than any other.
Here is the basic analysis using all state polls taken within the past month or, failing that, the most recent poll:
|95.8% probability of winning||4.2% probability of winning|
|Mean of 316 electoral votes||Mean of 222 electoral votes|
After 10,000 simulated elections, Obama wins 9,583 times and Romney wins 417 times (including the 34 ties). Obama receives (on average) 316 to Romney’s 222 electoral votes. Obama has a 95.8% probability of winning and Romney has a 4.2% probability of winning.
Here is the distribution of electoral votes [FAQ] from the simulations:
The long term trends in this race can be seen from a series of elections simulated for every week for the past nine months (08 Apr 2011 to 08 Jan 2012). Each week polls are used from the preceding 1 month (or, failing that, the single most recent poll—FAQ).
Over the past nine months, Obama has maintained a clear advantage over Romney. Aside from a brief spell in late November, Obama’s probability of winning has remained above 75% and typically higher. There does not appear to be any clear trend over time—pretty much steady state.
(You can see the maps and electoral vote distributions animated here.)
Back to the current time period, here are the ten most probable electoral vote outcomes for Obama:
- 312 electoral votes with a 1.63% probability
- 304 electoral votes with a 1.57% probability
- 310 electoral votes with a 1.56% probability
- 306 electoral votes with a 1.52% probability
- 319 electoral votes with a 1.48% probability
- 314 electoral votes with a 1.47% probability
- 313 electoral votes with a 1.45% probability
- 329 electoral votes with a 1.45% probability
- 298 electoral votes with a 1.42% probability
- 321 electoral votes with a 1.41% probability
After 10000 simulations:
- Obama wins 95.8%, Romney wins 4.2%.
- Average (SE) EC votes for Obama: 316.0 (28.1)
- Average (SE) EC votes for Romney: 222.0 (28.1)
- Median (95% CI) EC votes for Obama: 314 (264, 373)
- Median (95% CI) EC votes for Romney: 224 (165, 274)
Each column of this table shows the electoral vote total aggregated by different criteria for the probability of winning a state (Safe=100%, Strong=90%+, Leans=60%+, Weak=50%+):
|Threshold||Safe||+ Strong||+ Leans||+ Weak|
This table summarizes results by state. Click on the poll count to see the individual polls included for the state.
|4||8||Votes||polls||Votes||Obama||Romney||% wins||% wins|
* An older poll was used (i.e. no recent polls exist).
Details of the methods are given in the FAQ.
The most recent analysis in this match-ups can be found here.